scholarly journals TEXTURE CLUSTERING OF SATELLITE IMAGES USING SELF-ORGANIZING NEURAL NETWORK

2014 ◽  
pp. 15-21
Author(s):  
M. M. Lukashevich ◽  
R. Kh. Sadykhov
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Satellite Images ◽  
Image Data ◽  
Remote Sensing Image ◽  
Radar Images ◽  
Texture Information ◽  
Winner Takes All ◽  
Data Browsing ◽  
Self Organizing

The goal of this paper is to present a texture clustering system for remote sensing image data. Texture information is useful for image data browsing and retrieval. Authors present the results of self-organizing neural network design for solving the clustering task of gray scale remote sensing image data. The architecture of neural network and the learning algorithms for this network such as: algorithm WTA (Winner Takes All), algorithm CWTA (Winner Takes All with Conscience) and classic Kohonen algorithm WTM (Winner Takes Most - the Winner receives more) are considered. Some experimental results using textures of the Brodatz album, multi-spectral and radar images are also represented.

scholarly journals Scene classification of remote sensing image based on compound pruning

2021 ◽  
Vol 336 ◽  
pp. 06030
Author(s):  
Fengbing Jiang ◽  
Fang Li ◽  
Guoliang Yang
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Image Data ◽  
Remote Sensing Image ◽  
Convolution Neural Network ◽  
Model Parameters ◽  
Scene Classification ◽  
Remote Sensing Image Classification ◽  
And Storage

Convolution neural network for remote sensing image scene classification consumes a lot of time and storage space to train, test and save the model. In this paper, firstly, elastic variables are defined for convolution layer filter, and combined with filter elasticity and batch normalization scaling factor, a compound pruning method of convolution neural network is proposed. Only the superparameter of pruning rate needs to be adjusted during training. in the process of training, the performance of the model can be improved by means of transfer learning. In this paper, algorithm tests are carried out on NWPU-RESISC45 remote sensing image data to verify the effectiveness of the proposed method. According to the experimental results, the proposed method can not only effectively reduce the number of model parameters and computation, but also ensure the accuracy of the algorithm in remote sensing image classification.

scholarly journals A Small Target Detection Research Based on Dynamic Convolution Neural Network

2021 ◽  
Author(s):  
Chenshuai Bai ◽  
Kaijun Wu ◽  
Dicong Wang ◽  
Hong Li ◽  
Mingjun Yan ◽  
...  
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Convolutional Neural Network ◽  
Target Detection ◽  
Image Data ◽  
Remote Sensing Image ◽  
Convolution Neural Network ◽  
Small Target ◽  
Training Process ◽  
Small Target Detection

Abstract Because the detection effect of EfficientNet-YOLOv3 target detection algorithm is not very good, this paper proposes a small target detection research based on dynamic convolution neural network. Firstly, the dynamic convolutional neural network is introduced to replace the traditional convolutional neural network, which makes the algorithm model more robust; Secondly, in the training process, the optimization parameters are continuously adjusted to further strengthen the model structure; Finally, in order to prevent over fitting, the Learning Rate and Batch Size parameters are modified during the training process. remote sensing image The results of the proposed algorithm on RSOD remote sensing image data sets show that compared with the original EfficientNet-YOLOv3 algorithm, the (Average Precision, AP) value is increased by 1.93% and the (Log Average Miss Rate ,LAMR) value is reduced by 0.0500; The results of the proposed algorithm on TGRS-HRRSD remote sensing image data set show that compared with the original EfficientNet-YOLOv3 algorithm, the mAP value is increased by 0.07% and the mLAMR value is reduced by 0.0007.

scholarly journals A Small Object Detection Research Based on Dynamic Convolution Neural Network

2021 ◽  
Author(s):  
Chenshuai Bai ◽  
Kaijun Wu ◽  
Dicong Wang ◽  
Mingjun Yan
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Object Detection ◽  
Image Data ◽  
Remote Sensing Image ◽  
Data Sets ◽  
Small Object ◽  
Training Process ◽  
The Mean ◽  
Small Object Detection

Abstract In view of the fact that the detection effect of EfficientNet-YOLOv3 object detection algorithm is not very good, this paper proposes a small object detection research based on dynamic convolution neural network. Firstly, the dynamic convolutional neural network is introduced to replace the traditional, which makes the algorithm model more robust; secondly, the optimization parameters are continuously adjusted in the training process to further strengthen the model structure; finally, the Learning Rate and Batch Size parameters are modified during the training process in order to prevent overfitting. In order to verify the effectiveness of the proposed algorithm, RSOD and TGRS-HRRSD remote sensing image data sets are used to test the effect. The results of the proposed algorithm on RSOD remote sensing image data sets show that compared with the original EfficientNet-YOLOv3 algorithm, the mean Average Precision (mAP) value is increased by 1.93% and the mean Log Average Miss Rate (mLAMR) value is reduced by 0.0500; The results of the proposed algorithm on TGRS-HRRSD remote sensing image data set show that compared with the original EfficientNet-YOLOv3 algorithm, the mAP value is increased by 0.07% and the mLAMR value is reduced by 0.0007.

Research on Remote Sensing Image Target Recognition Based on Deep Convolution Neural Network

2019 ◽  
Vol 34 (05) ◽  
pp. 2054015 ◽  
Author(s):  
Xiaofeng Han ◽  
Tao Jiang ◽  
Zifei Zhao ◽  
Zhongteng Lei
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Convolutional Neural Network ◽  
Target Recognition ◽  
Image Data ◽  
Network Models ◽  
Remote Sensing Image ◽  
Remote Sensing Images ◽  
Neural Network Models ◽  
Current Mass

Target recognition is an important application in the time of high-resolution remote sensing images. However, the traditional target recognition method has the characteristics of artificial design, and the generalization ability is not strong, which makes it difficult to meet the requirement of the current mass data. Therefore, it is urgent to explore new methods for feature extraction and target recognition and location in remote sensing images. Convolutional neural network in deep learning can extract representative and discriminative multi-level features of typical features from images, so it can be used for multi-target recognition of remote sensing big data in complex scenes. In this study, NWPU VHR-10 data was selected, 50% was used for training, and the remainder was used for verification. The target recognition effects of two kinds of convolutional neural network models, Faster R-CNN and SSD, were studied and compared, and the mean average precision (mAP) was used for evaluation. The evaluation results show that the Faster R-CNN has three categories with an accuracy of more than 80%, and the SSD has seven categories with an accuracy of more than 80%, all of which show good results. The SSD model is particularly prominent in running time and recognition results, which proves convolutional neural networks have broad application prospects in the target recognition of remote sensing image data.

MILL: Channel Attention–based Deep Multiple Instance Learning for Landslide Recognition

2021 ◽  
Vol 17 (2s) ◽  
pp. 1-11
Author(s):  
Xiaochuan Tang ◽  
Mingzhe Liu ◽  
Hao Zhong ◽  
Yuanzhen Ju ◽  
Weile Li ◽  
...  
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Large Scale ◽  
Remote Sensing Image ◽  
Disaster Risk ◽  
Multiple Instance Learning ◽  
Remote Sensing Images ◽  
Loess Area ◽  
Remote Sensing Image Classification ◽  
Natural Disaster Risk

Landslide recognition is widely used in natural disaster risk management. Traditional landslide recognition is mainly conducted by geologists, which is accurate but inefficient. This article introduces multiple instance learning (MIL) to perform automatic landslide recognition. An end-to-end deep convolutional neural network is proposed, referred to as Multiple Instance Learning–based Landslide classification (MILL). First, MILL uses a large-scale remote sensing image classification dataset to build pre-train networks for landslide feature extraction. Second, MILL extracts instances and assign instance labels without pixel-level annotations. Third, MILL uses a new channel attention–based MIL pooling function to map instance-level labels to bag-level label. We apply MIL to detect landslides in a loess area. Experimental results demonstrate that MILL is effective in identifying landslides in remote sensing images.

scholarly journals Archaeological Surveying of Subarctic and Arctic Landscapes: Comparing the Performance of Airborne Laser Scanning and Remote Sensing Image Data

2021 ◽  
Vol 13 (4) ◽  
pp. 1917
Author(s):  
Alma Elizabeth Thuestad ◽  
Ole Risbøl ◽  
Jan Ingolf Kleppe ◽  
Stine Barlindhaug ◽  
Elin Rose Myrvoll
Keyword(s):  
Remote Sensing ◽  
Laser Scanning ◽  
Digital Terrain Model ◽  
Image Data ◽  
Remote Sensing Image ◽  
Airborne Laser Scanning ◽  
Aerial Images ◽  
Detection Rates ◽  
Airborne Laser ◽  
Pros And Cons

What can remote sensing contribute to archaeological surveying in subarctic and arctic landscapes? The pros and cons of remote sensing data vary as do areas of utilization and methodological approaches. We assessed the applicability of remote sensing for archaeological surveying of northern landscapes using airborne laser scanning (LiDAR) and satellite and aerial images to map archaeological features as a basis for (a) assessing the pros and cons of the different approaches and (b) assessing the potential detection rate of remote sensing. Interpretation of images and a LiDAR-based bare-earth digital terrain model (DTM) was based on visual analyses aided by processing and visualizing techniques. 368 features were identified in the aerial images, 437 in the satellite images and 1186 in the DTM. LiDAR yielded the better result, especially for hunting pits. Image data proved suitable for dwellings and settlement sites. Feature characteristics proved a key factor for detectability, both in LiDAR and image data. This study has shown that LiDAR and remote sensing image data are highly applicable for archaeological surveying in northern landscapes. It showed that a multi-sensor approach contributes to high detection rates. Our results have improved the inventory of archaeological sites in a non-destructive and minimally invasive manner.

Sign in / Sign up

Export Citation Format

Share Document